Research On Power Control Modeling Of Wind Turbine Based On Data Statistical Analysis

At the moment,the fundamental challenge to be tackled for any country’s economic development and sustainable development is the energy crisis.Wind energy is pure and renewable sources,and abundant wind energy reserves in the china,it is imperative to vigorously develop wind power generation technology.Demestic research on wind turbine system desigin still needs to be consummate.In order to make the wind turbine design accord with the reality environment,Therefore,this paper takes SCADA data modeling of wind turbines as the starting point,and establishes Bladed model the same as real wind turbines.The particular analysis content of this paper is as follows.Firstly,describe the structure of wind turbine,and analyze and introduce the characteristics and power control strategy of the unit,which provides fundamental basis for the follow-up wind turbine moduling and controller design.Secondly,the data from SCADA system of a specific 1.5MW wind turbine unit in a wind farm is statistically evaluated,and the operation of the wind turbine is separated into four sections based on wind speed.Meanwhile,preprocess the datas,this is,renovating the anomalous datas and padding in the defect datas,to guarantee the effectiveness of the data.Then,the least square method and fuzzy weighted least square method are used to model the wind turbine running state under maximum wind energy acquire stage and constant power stage respectively,and an evaluation quate of the model is put forward.Analyze the strengths and weaknesses of various moulds and the best mould is selected,and the optimal model is verified which is convenient for comparative analysis the simulation results with the GH-Bladed model.Then,Bladed software is used to digitally model each part of the 1.5MW real wind turbine.Make the Bladed wind file based on the real wind speed,and the deviation degree is analyzed by euclide distance to ensure that the wind speed of the Bladed model is the same as that of the real wind turbine in the natural environment.Finally,through the analysis of SCADA data,the Bladed power controller is designed.The optimal control method and look-up table method are compared and analyzed,the look-up table approach is chosen as the torque controller’s control method.SCADA data is used to developed the pitch controller’s system identification model under constant power settings,and the initial PI parameters of the pitch controller for the system identification model are determined using a genetic algorithm.Bladed model simulation,the operaing condition of maximum wind energy acquire parameters such as generator speed,torque,power and constant gain operating parameters such as pitch angle,power and compared with the SCADA data,the two working situations’ running states anddata models are compared and assessed.The result clear that within the permitable error range,the coverage of parameter is more than 95%,and the data model is consistent with the simulation running state of the two operating circumstances.The Bladed model can be equivalent to a real wind turbine.